2.4. TFIIH Recruitment and Functions in TC-NER
TC-NER only repairs damage on the transcribed strand of active genes and is considered more efficient than GG-NER (Hu et al., 2015). A key difference that distinguishes TC-NER from GG-NER is the presence of damage-stalled RNA Pol II that serves as the signal for TC-NER initiation (Lainé & Egly, 2006). The first protein responding to Pol II stalling is Cockayne syndrome B (CSB), a SWI2-SNF2 type ATPase (Selby & Sancar, 1997). CSB normally binds to DNA upstream of Pol II to promote transcription elongation (Kokic et al., 2021; Xu et al., 2017). Upon transcription stalling, CSB quickly moves to Pol II and functions in recruiting downstream TC-NER proteins, including Cockayne syndrome A (CSA) (van der Weegen et al., 2020), a component of an E3 ubiquitin ligase complex (Groisman et al., 2003). CSA can ubiquitylate CSB as well as the stalled Pol II (Groisman et al., 2006; Nakazawa et al., 2020). CSA also physically interacts with UV-stimulated scaffold protein A (UVSSA) (van der Weegen et al., 2020), another important TC-NER protein.
One mechanism for TFIIH recruitment in TC-NER is through its physical interaction with UVSSA (Okuda et al., 2017; van der Weegen et al., 2020) (Figure 3 ). In this regard, it has been shown that UVSSA also interacts with the PH domain of TFIIH subunit p62 (Okuda et al., 2017), in a way similar to the interaction between XPC and TFIIH in GG-NER. Another mechanism for TFIIH recruitment is via Pol II ubiquitylation. Recruitment of CSA to the stalled Pol II leads to mono-ubiquitylation of the largest Pol II subunit, Rpb1, at Lys1268 (Nakazawa et al., 2020; Tufegdžić Vidaković et al., 2020). Interestingly, Rpb1-Lys1268 ubiquitylation enhances the association of the TFIIH core complex with the stalled Pol II, and this mechanism appears to involve ubiquitylated UVSSA at Lys414 (Nakazawa et al., 2020). An additional TC-NER factor that may participate in TFIIH recruitment to stalled Pol II is ELOF1. It was suggested that ELOF1, a conserved elongation factor, interacts with both Pol II and the CRL4CSA E3 ligase, and positions CRL4CSA for Pol II ubiquitylation at the Rpb1-Lys1268 residue(van der Weegen et al., 2021). As Pol II ubiquitylation increases Pol II-TFIIH interaction (Nakazawa et al., 2020), ELOF1 likely facilitates this process by enhancing Pol II ubiquitylation.
Despite TFIIH’s roles in DNA unwinding and damage verification in GG-NER, how TFIIH stimulates TC-NER is much less understood. It is generally assumed that TFIIH plays identical roles in the two NER subpathways and there is some evidence supporting this hypothesis. For example, it has been shown that a helicase-dead XPD mutant abolishes both subpathways in yeast (Duan et al., 2020). However, it is also important to note that TC-NER significantly differs from GG-NER in that the two DNA strands are pre-melted in a transcription bubble by RNA Pol II (Figure 3 ). When RNA Pol II is stalled by the damage, it is conceivable that TFIIH may not need to unwind the two strands from scratch, instead, it is possible that TFIIH may just need to extend the transcription bubble to ~30 nt for the formation of NER pre-incision complex. Consistent with this notion, clinical data have shown that mutations in XPD, the major helicase responsible for DNA unwinding, are mainly associated with the skin cancer-prone disease, xeroderma pigmentosum (XP), which is generally considered to be caused by GG-NER defects (Coin et al., 1998; Lehmann, 2001). Only a small number of XPD mutations are associated with the severe symptom of XP in combination with the TC-NER disease, Cockayne syndrome (CS) (Lehmann, 2001; Rapin et al., 2000).
One possible explanation for the clinical observations is that the XPD mutations in most patients may retain partial helicase activity that is strong enough to increase the bubble size using the pre-melted DNA in TC-NER. However, the attenuated helicase activity may not be enough for generating a repair bubble in GG-NER on an almost fully annealed DNA double helix. More detailed DNA repair studies in different XPD mutant cells (e.g., XP-only or XP plus CS) may help us understand the underlying mechanism for different XPD symptoms and delineate the exact roles of TFIIH in the two subpathways. Furthermore, to what extent XPD’s damage verification function is required for TC-NER is also up for debate; because RNA Pol II stalling should already be a stringent mechanism to verify the presence of DNA damage. Whether TC-NER needs both Pol II stalling and TFIIH to verify damage presence needs more experimental analysis.
It is also still not fully understood if RNA Pol II is evicted from the DNA to make way for the TFIIH repair complex, along with other NER factors, or if it simply backtracks along the DNA in the transcription bubble. There are a number of theories about what could be happening, but each raises its own questions. If RNA Pol II dissociates from the DNA, how is it recruited back? Does it retain the transcript in progress, or does it need to start at the promoter region again? If RNA Pol II is backtracked, what is the mechanism promoting Pol II backtracking along the DNA? Considering TFIIH’s DNA helicase function, future studies should also test a potential role for TFIIH in aiding Pol II dissociation from DNA or backtracking in TC-NER.